Mitochondrial impairment and downregulation of Drp1 phosphorylation underlie the antiproliferative and proapoptotic effects of alantolactone on oral squamous cell carcinoma cells

The human OSCC cells, including CAL27 and SCC9 cells, were from Wuhan University. CAL27 cells were cultured in DMEM and SCC9 cells were cultured in F-12K medium. Cells were cultured at 37 °C with 5% CO₂. When the cells achieved 80% confluency, they were passaged and digested with trypsin. Cultured cells were treated with ALT of different concentrations and duration time. For experiments with NAC, the cells were pre-treated with 5 mM NAC for 2 h, according to the previous studies [28, 29].

Drp1 overexpressed plasmids were transfected into cells to explore the role of Drp1 in OSCC progression. The cells were inoculated into a 6-well plate and transfected with control and Drp1 overexpressed plasmids. All transfections were performed using Lipofectamine 3000 (L3000008, Invitrogen, USA) according to the manufacturer’s instructions [30]. Transfection efficiency was confirmed by western blot.

We adopted the Methyl thiazolyl tetrazolium (MTT) assay to assess the cell viability. Cells were seeded in a 96-well plate at a density of 4500 cells/well in a volume of 100 μL/well. After incubating with MTT, dimethyl sulfoxide was added to the cells to dissolve the formazan crystals. The spectrophotometric absorbance was measured at 490 nm. The average absorbance of the control DMSO group was set at 100%, and the absorbance ratio reflected the cell viability.

Cells were seeded in a 6-well plate at 2 × 10⁵ cells/well density and cultured under different conditions. Total proteins were collected using RIPA buffer (P0013, Beyotime, China), the concentrations of which were determined using the Bradford protein assay kit (23236, Thermo Fisher Scientific, USA). During the electrophoresis, we loaded 60–80 µg proteins in each slot in the SDS gel. Protein gels were blotted using the Trans-Blot Turbo transfer apparatus and PVDF Midi transfer packs (Bio-Rad). After blocking with 5% non-fat milk in TBST, blots were incubated with the primary antibodies with anti-p-Drp1 (1:1000), anti-Drp1 (1: 1000), and anti-GAPDH (1:20,000). After further incubation with secondary antibodies, blots were developed with ECL substrate (32209, Thermo Fisher Scientific, USA). NIH Image J software (https://​imagej.​nih.​gov/​ij/​) detected the immunoreactive band relative to the optical density.

2,7-Dichlorofluorescein diacetate (DCFH-DA) (S0033, Beyotime, China) was used to evaluate the intracellular ROS production. 3-amino,4-aminomethyl-2′,7′-fluorescein diacetate (DAF-FM DA) (S0019, Beyotime, China) was used to detect intracellular RNS level [31]. After treatment, the cells were stained with either DCFH-DA or DAF-FM DA for 20 min and rinsed with PBS. Images were taken by fluorescence microscope (Nikon, Japan) and evaluated by NIH Image J software (https://​imagej.​nih.​gov/​ij/​).

Cells were incubated with MitoSOX red to detect mitochondrial ROS (mtROS). Additionally, mitochondrial membrane potential (MMP) was evaluated with the tetramethylrhodamine methyl ester dye (TMRM). We detected the fluorescence signals of MitoSOX (M36006, Thrmofisher, USA), TMRM (I34361, Thrmofisher, USA), and MitoGreen (M36008, Thrmofisher, USA) at the excitation/emission wavelengths of 357/410 nm, 548/574 nm, and 491/513 nm. Fluorescence signals were measured using NIH Image J software.

Crystal violet (0.5% in 25% methanol) was used to stain the cells under different treatment conditions. A microscope (Olympus, Japan) was used to record the images. The colony formation areas were evaluated by Image J software (https://​imagej.​nih.​gov/​ij/​).

An Annexin V-FITC/PI (C1062, Beyotime, China) apoptosis detection kit was adopted to detect the apoptotic rate. After trypsin digestion without ethylene diamine tetraacetic acid, the cells were collected by centrifugation and re-suspended. Annexin V-FITC and PI were added to the cell suspension in an ice bath. These staining conditions were examined by flow cytometer (CytoFLEX; Beckman Coulter, USA).

ATP is primarily produced through oxidative phosphorylation in mitochondria and thus represents an essential indicator of mitochondrial function [32]. The CellTiter-Glo Assay evaluated ATP production. 100 µL of ATP detection buffer was added into each well. The luminescence was measured after 10 min of incubation, and the ATP level was normalized based on protein concentration.

The specimens were formalin-fixed, paraffin-embedded, and cut into sheets of 4 µm. Gradient alcohol dehydrated the tissue slides after they were dewaxed with xylene. The sections were added with diluted primary antibody against p-Drp1 (1:200), Drp1 (1:200), TOM20 (1:200), and VDAC1 (1:200). Nuclei were counterstained with hematoxylin and then sealed in neutral rubber. In the final step, the sections were viewed under a light microscope.

Wenzhou Medical University’s Research Ethics Committee approved protocol KY2022-R156 for this clinical-laboratory study. From October 2020 to December 2021, samples were obtained from 20 paired cancer tissues and adjacent tissues from patients at the Department of Oral and Maxillofacial Surgery, the First Affiliated Hospital of Wenzhou Medical University. Inclusion criteria were: (1) Patients complied with the Head and Neck Cancers Clinical Practice Guidelines [33]. (2) None of the patients had previously been treated with chemotherapy or radiotherapy. (3) The patients have no other malignancy or autoimmune diseases.

The RNA-seq transcriptome data of 500 head and neck squamous cell carcinoma tissues, 44 adjacent normal tissues and corresponding clinical data were obtained from the Cancer Genome Atlas (TCGA) (http://​cancergenome.​nih.​gov/​) database. MitoCarta2.0 documented the biological pathways and gene sets related to mitochondrial metabolism and thus was used in our study. Single-sample gene set enrichment analysis was used to calculate the infiltrating scores of mitochondrial-related metabolic pathways based on the above information. A protein–protein interaction (PPI) network was constructed using the Interacting Gene Retrieval Search tool (https://​string-db.​org/​) to evaluate the interactions of mitochondrial metabolic pathways-related genes (MMPRGs). MMPRGs with |log Fold Change| > 0.5 and an adjusted p-value < 0.05 were identified as differentially expressed MMPRGs. Genes were further evaluated for prognostic significance with univariate cox regression analysis.

Data analysis was performed with GraphPad Prism 8.0. Analyses of variance were performed to evaluate the data, expressing it as mean ± SD. Statistics were significant when p < 0.05. Six photographic fields were detected from all the figures and used for analysis. Each experiment was repeated three times. All bioinformatic analyses were conducted by packages and appropriate scripts in R (version 4.2.1) (http://​www.​R-project.​org).

Alantolactone (ALT) is a natural sesquiterpene lactone isolated from Inula helenium (Fig. 1A). We examined the effects of ALT on various types of cancer cells, including non-small cell lung cancer cell line PC-9, human LoVo colon cancer line, human glioblastoma cell lines U251, and the OSCC cell lines with CAL27 and SCC9 cells. The results showed that the OSCC cells were more sensitive to ALT than the other cancer subtypes (Fig. 1B), suggesting the strong inhibitory effect of ALT on OSCC. We further investigated the effects of ALT on CAL27 and SCC9 cell lines at different time intervals. ALT reduced the cell viability in a time-dependent way, with lower viability early at 12 h than the control group (Fig. 1C, D). In addition, ALT-treated cells formed smaller and fewer clones than their control counterparts in colony outgrowth assays (Fig. 1E–G). Taken together, ALT inhibited cell proliferation and colony formation abilities, which are essential for the formation of OSCC. We further explored the impact of ALT on the cell apoptosis rate of CAL27 and SCC9 cells. ALT increased the apoptotic rate of CAL27 and SCC9 cells, respectively, from 18.52 to 66% and from 2.81 to 49.75% (Fig. 1H–K).

To check whether ALT affects CAL27 and SCC9 cells via the regulation of OS, we measured the cellular total ROS level with the fluorescent probe DCFH-DA. ROS levels increased dose-dependently after ALT treatment (Fig. 2A–C). We further measured the RNS level as an additional readout of OS through DAF-FM DA. ALT dose-dependently increased the RNS levels in CAL27 and SCC9 cells (Fig. 2D–F). We further used NAC, a ROS scavenger, to identify whether ROS mediates the anticancer property of ALT. Compared with the cells only treated with ALT, a combination of NAC pretreatment and ALT treatment resulted in significantly lower ROS production in CAL27 and SCC9 cells, as confirmed both by DCFH-DA staining and flow cytometry (Fig. 2G–I). Quantitative analysis of MTT and colony outgrowth assay also showed that cells cotreated with NAC and ALT presented a higher proliferation rate than the lean ALT treatment group (Fig. 3A–F). Furthermore, NAC treatments reduced ALT-induced cell apoptosis (Fig. 3G–J). These results strongly substantiated that ROS regulation played crucial roles in the inhibitory effects of ALT on CAL27 and SCC9 cells.

Regarding the close relationship between ROS and mitochondria, it is necessary to investigate whether mitochondria play a role in the inhibitory effect of ALT on OSCC. ALT treatment decreased the ATP levels in CAL27 and SCC9 cells (Fig. 4A, B). To further explore whether ALT-induced ROS production was derived from mitochondria, we used MitoSOX to indicate the mtROS production. The results showed that after ALT treatment, mtROS production in CAL27 and SCC9 cells was vitally increased (Fig. 4C–F). Besides, TMRM staining indicated that ALT markedly decreased MMP (Fig. 4G–J). These results suggested that ALT induced remarkable mitochondrial impairment, as indicated by increased mtROS level and MMP loss.

We further applied NAC to probe the specific role of ROS in regulating ALT-elicited mitochondrial impairment. NAC dramatically reversed the ALT-elicited ATP depletion in CAL27 and SCC9 cells (Fig. 5A, B). Furthermore, ALT increased the mtROS production, which was reversed by NAC cotreatment. These results suggested that ALT-induced intracellular ROS production was mainly derived from mitochondria (Fig. 5C–F). Moreover, Cells treated with NAC and ALT had significantly higher MMP levels than those only treated with ALT, indicating that NAC reversed the MMP loss induced by ALT (Fig. 5G–J). These experiments revealed that ROS-related mitochondrial abnormalities played a crucial role in the ALT-elicited CAL27 and SCC9 cell apoptosis.

To further explore the possible mitochondrial pathways involved in OSCC progression, we evaluated the gene expression patterns across diverse human OSCC and normal tissues with the TCGA database. Results showed that compared to the control group, mtDNA maintenance and mitochondrial dynamics were the two most differentially expressed pathways increased in the cancer tissues. Moreover, the pathways, such as amino acid metabolism and OXPHOS subunits, decreased significantly in the OSCC cancer tissues, which needs further exploration (Fig. 6A). Regarding the remarkable difference in mtDNA maintenance and mitochondrial dynamics between the normal and OSCC patients, we next focused on the possible mitochondrial regulating proteins involved in OSCC progression. Based on the STRING database, a PPI network of differentially expressed mitochondrial and mtDNA genes were constructed (Fig. 6B), and the top 20 proteins were obtained (Fig. 6C). Among the top 20 proteins, Drp1 (alias DNM1L), TOM20 (alias TOMM20), and VDAC1 differed significantly in cancer progression. Therefore, we further focused on the role of Drp1, TOM20, and VDAC1 in OSCC development. We collected cancer and normal tissues from 20 OSCC patients during surgery. ATP levels were strikingly higher in the cancer tissues than in the normal tissues (Fig. 6D). This result indicated that mitochondria played a crucial role in OSCC, as mitochondria are the primary source of ATP production. In OSCC cancer tissues, p-Drp1, Drp1, and TOM20 expressions increased significantly compared with normal tissues. Furthermore, the p-Drp1/Drp1 level increased significantly in the cancer samples (Additional file 1: Fig. S1). By contrast, the VDAC1 level was lower in OSCC cancer tissues than in normal tissues (Fig. 6E). Also, TCGA data revealed that OSCC patients with low Drp1 expression in cancer tissues had a higher overall survival rate than those with high Drp1 expression (p = 0.008) (Fig. 6F). More importantly, we explored whether Drp1 regulation participated in the inhibitory effects of ALT on OSCC cells. The results revealed that ALT dose-dependently downregulated the Drp1 phosphorylation level (Fig. 6G). We further investigated to explore the association between ROS generation and Drp1 dephosphorylation in the inhibitory effect of ALT on OSCC. The results confirmed that the ALT-induced decline of Drp1 dephosphorylation was reversed by NAC (Fig. 6H–J), indicating that ALT inhibited OSCC cell apoptosis by suppressing ROS-related Drp1 dephosphorylation.

Drp1 overexpression abolished the reduced Drp1 phosphorylation by ALT (Fig. 7A–D) and promoted the cell viability reduced by ALT (Fig. 7E, F). Drp1 overexpression also reversed the mitochondrial dysfunction induced by ALT, with increased ATP level (Fig. 8A, B), decreased ROS production (Fig. 8C–F), and increased mitochondrial membrane potential (Fig. 8G–J) compared to the control group.